Infectious bursal disease (“IBD”), an infectious disease among young chickens, was first recognized in 1957 in Gumboro, Del., USA, and formally documented by Cosgrove (Cosgrove, 1962; Lasher and Shane, 1994). As a result, the disease is often referred to as “Gumboro”. Not long after IBD was first reported, it was recognized in poultry populations throughout the world (Lasher and Shane, 1994). IBD is caused by a virus (“IBDV”) classified as a Birnavirus (Dobos et al., 1979).
Two different IBDV serotypes exist: serotypes I and II (Jackwood et al., 1982; McFerran et al., 1980). Isolates belonging to serotype I are highly pathogenic for chickens. Serotype II isolates, which are mainly recovered from turkeys, have not been reported to induce clinical signs in chickens and are regarded as apathogenic (Ismail et al., 1988). Infectious bursal disease or Gumboro is a highly contagious disease for young chickens and is responsible for severe losses in the poultry industry. In birds surviving an acute infection, lymphoid cells in the bursa of Fabricius are destroyed, resulting in B-cell-dependent immunodeficiency. This causes increased susceptibility to disease caused by otherwise harmless agents. A central role in the pathogenesis of Gumboro is played by the bursa, which is representing the target organ of the virus.
IBDV infections were initially recognized by whitish or watery diarrhea, anorexia, depression, trembling, weakness, and death. This clinical IBD was generally seen in birds between three and eight weeks of age. The course of the disease runs approximately 10 days in a flock. Mortality usually ranges from 0–30 percent. Field reports suggest that leghorns are more susceptible to IBDV than broiler-type chickens. Subclinical IBD was later recognized and is generally considered a greater problem in commercial poultry than the clinical disease. It is generally seen in birds less than three weeks of age. This early infection results in a B-lymphocyte depletion of the bursa of Fabricius. The bird is immunologically crippled and unable to respond fully to vaccinations or field infections. In susceptible chickens, damage caused by IBDV can be seen within two to three days after exposure to virulent virus. Initially, the bursa swells (3 days post-exposure) with edema and hemorrhages and then begins to show atrophy (7–10 days). IBD virus is especially cytopathic to certain B-lymphocytes. The highest concentration of these specific B-lymphocytes is found in the bursa. Destruction of the B-lymphocytes by IBD field virus may result in an incomplete seeding of these cells in secondary lymphoid tissue. As a result of the depletion of B-lymphocytes, surviving birds are immunocompromised during the remainder of their lifetime.
IBDV is found worldwide, and IBDV-specific antibodies have even been found in Antarctic penguins (Gardner et al., 1997). The prevalence of clinical IBD is relatively low compared to the prevalence of subclinical IBD. IBDV is very resistant to common disinfectants and has been found in lesser mealworms, mites, and mosquitoes. These facts correlate with field experience of reoccurring IBD problems on a farm, despite clean-up efforts. Infection with IBDV results in a strong antibody response against IBD, which is capable of neutralizing this virus. Most likely as a result of vaccination, antigenic variant isolates of serotype I were isolated in the Delaware area (USA). These isolates have been shown to cause bursa atrophy in as little as three days post-infection without inflammation of the bursa. Despite their change in antigenicity, these antigenic variants do not form a distinct serotype. After the occurrence of antigenic variant IBDV isolates in the USA, the poultry industry in European countries was hit by outbreaks of IBD caused by a very virulent serotype I IBDV (vvIBDV) (Berg et al., 1991; Chettle et al., 1989; Kouwenhoven and Van den Bos, 1995). These very virulent field isolates were capable of establishing themselves in the face of high levels of maternal antibodies which normally were protective. These vvIBDV cause more severe clinical signs during an outbreak and are now found globally (e.g., Europe, Japan, Israel and Asia).
IBDV belongs to the family of Birna viruses which include Infectious Bursal Disease Virus (IBDV) isolated from chickens, Infectious Pancreatic Necrosis Virus (IPNV) isolated from Fish, Drosophila X Virus (DXV) isolated from fruit fly, and Tellina Virus (TV) and Oyster Virus (OV) both isolated from bivalve molluscs (Dobos et al., 1979). Birna viruses have a dsRNA genome which is divided over two genome segments (the A- and B-segment). The A-segment (3.3 kbp) contains two partly overlapping open reading frames (ORFs). The first, smallest ORF encodes the non-structural Viral Protein 5 (VP5, 17 kDa). The second ORF encodes a polyprotein (1012 amino acid, 110 kDa), which is autocatalytically cleaved. The exact positions of these cleavage sites are unknown.
From SDS-Page analysis of in vitro translated IBDV RNA, it is known that the polyprotein is rapidly cleaved into three proteins: pVP2 (48 kDa), VP4 (29 kDa) and VP3 (33 kDa). During in vivo virus maturation, pVP2 is processed into VP2 (38 kDa), probably resulting form site-specific cleavage of the pVP2 by a host cell-encoded protease (Kibenge et al., 1997). VP2 and VP3 are the two proteins that constitute the single shell of the virion. The B-segment (2.9 kbp) contains one large ORF, encoding the 91 kDa VP1 protein. This protein contains a consensus RNA-dependent RNA polymerase motive (Bruenn, 1991).
Furthermore, this protein has been reported to be linked to the 5′-ends of the genomic RNA segments (Viral Protein genome-linked, VPg). The nucleotide sequence of internal parts of a large number of IBDV isolates of classical, antigenic variant or very virulent origin has been determined and deposited in several databases such as GenBank. Furthermore, Mundt and Muller (Mundt and Muller, 1995) have determined the 5′- and 3′-termini of several IBDV isolates (CU-1, CU-1M, P-2 and 23/82), and by combining the internal and terminal sequences, Mundt and Muller established the complete nucleotide sequence of a serotype I A-segment (3261 bp) and B-segment (2827 bp). This provided the way to generate an infectious (recombinant) copy (rIBDV) of IBDV serotype I, by knowing the complete sequence dsRNA sequence of IBDV genome and by using one of several methods to generate infectious copy virus (see, for example, Boyer et al., Virology 198:415–426, 1994). Mundt and Vakharia indeed produced infectious rIBDV serotype I from cDNA (Mundt and Vakharia, 1996). Full-length cDNA of a serotype I IBDV, preceded by a T7 promoter, was thereby used as a template for T7 RNA polymerase using a method described by Weiland and Dreher (Weiland and Dreher, 1989). The in vitro-generated mRNA, containing a cap-structure at its 5′-end, was subsequently transfected into eukaryotic cells (VERO cells) using a liposome formulation (Lipofectin, GibcoBRL). The supernatant of the transfected cells contained infectious rIBDV after incubation during 36 h in the CO2 incubator at 37° C. (Mundt and Vakharia, 1996; (WO 98/09646)). In addition, Lim et al. introduced two amino acid mutations (D279N and A284T) into the cDNA of vvIBDV isolate HK46 (Lim et al., 1999). These mutations were most probably based on data of Yamaguchi et al. (Yamaguchi et al., 1996), which showed that these specific mutations were found in two independent experiments in which very virulent IBDV isolates lost their very virulent character by adaptation and growth on primary CEF cells. Lim et al. obtained an rIBDV isolate which possessed the phenotype of a CEF-culture adapted isolate, i.e., an rIBDV isolate which can be propagated, i.e., is able to infect, multiply and be released for further replication, in vvIBDV non-permissive cells such as CEF cells. Noteworthy, Lim et al. were unable to produce an infectious vvIBDV isolate using the unmodified cDNA of the HK46 isolate (Lim et al., 1999). Furthermore, although cDNA of IBDV can be used to produce infectious IBDV, the exact mechanism of replication has not been elucidated yet. Data exist which are in support of a semi-conservative genome replication model for Birnaviradae (Bernard, 1980; Mertens et al., 1982).
Now and then, IBDV variants are detected in the field or are created in cell cultures in the laboratory (Muller, 1987) that are genetic re-assortments of serotype I and II strains of IBDV, in that they contain one genomic segment derived from the one serotype and another segment derived from the other serotype. Such segment re-assorted (srIBDV) strains (also called chimeric IBDV) not only occur in nature, but have recently been generated from cDNA as a well, by Vahkaria and Mundt (WO 98/09646).
Vaccination using attenuated field isolates worked sufficiently well until antigenic derivatives were found in the Delaware region of the USA starting in 1985 (isolates Del A, D, G and E) (Snyder, 1990). These field isolates were missing an important virus-neutralizing epitope. The change of this epitope is characterized by the lack of binding of the virus-neutralizing monoclonal antibody (Mab) B69 (Snyder et al., 1988a). The antibodies induced by vaccination with classical IBDV vaccines appeared to be less protective against these antigenic IBDV variants. Inactivated vaccines based upon antigenic IBDV variants were subsequently produced and were found to protect effectively against these antigenic variants of IBDV. After the Delaware variant, a second antigenic variant IBDV was isolated. This variant was recovered from the Delmarva region (USA) and was referred to as the GLS variant. The GLS variant is characterized by the absence of epitopes for both the virus-neutralizing Mab B69 and R63 (Snyder et al., 1988b). After identifying these antigenic variants, a large survey was performed within the USA by using a panel of nine Mabs against IBDV. This survey yielded an additional antigenic variant: the DS326 variant. This antigenic variant is characterized by the absence of epitopes for Mab 179 and BK44, in addition to those for Mabs B69 and R63 (Snyder, 1990). No further reports of antigenic variants have been published in the USA or in other parts of the world. Whether this is due to non-existence of new variant IBDV isolates or whether new antigenic variants just have not been detected due to the lack of extensive surveys or the lack of discriminating monoclonal antibodies is unclear.
The nucleotide sequence of the polyprotein-encoding part of the A-segment of the Del, the GLS and the DS326 antigenic variant IBDV isolates has been determined (Vakharia et al., 1994). Most of the amino acid changes were found in a specific region of the VP2 protein, the so-called hypervariable region. Furthermore, it was found that the epitopes which are capable of inducing neutralizing antibodies are conformation dependent and are clustered in the hypervariable region. This region consists of a domain with a high hydrophobicity index (amino acids 224 to 314 of pVP2, corresponding with amino acids 224 to 314 of the polyprotein) which is flanked by two small hydrophilic regions, each spanning about 14 amino acids (Vakharia et al., 1994; Heine et al., 1991). Amino acid substitution both within the hydrophobic region and within the hydrophilic regions might be involved in the antigenic variant character of these isolates.
After the problems caused by the antigenic variant IBDV isolates in the USA, the poultry industry in Europe was affected by very virulent IBDV (vvIBDV) isolates (Berg et al., 1991; Chettle et al., 1989). The vvIBDV isolates cause more severe clinical signs during an outbreak and are able to break through levels of antibodies which are protective against classical IBDV isolates. The molecular determinants which distinguish vvIBDV from classical IBDV isolates are not exactly known. It is known, however, that the pathogenicity of cell culture adapted very virulent IBDV isolates is severely reduced, compared with the non-adapted parental isolates (Yamaguchi et al., 1996). The correlation between CEF-adaptation and loss of the very virulent phenotype is likely to be due to the change in target cell tropism of the adapted virus. This change in cell tropism may be due to the loss of bursa cell receptor binding capability of the cell culture adapted very virulent IBDV isolate. Another possibility is that the cell culture adapted very virulent IBDV isolate is able to infect non-bursa cells, resulting in large reduction of IBDV load in the primary target cells (bursa cells). From the published results (Yamaguchi et al., 1996), it is clear that a recombinant IBDV (rIBDV) which is based upon the cDNA of a cell culture adapted very virulent isolate will never yield a vaccine which meets the demands of being able to break through high levels of maternal antibodies and induce a high enough immune response.
No specific antibodies that exclusively recognize the vvIBDV isolates have been described yet (Eterradossi et al., 1997). The lack of discriminating antibodies makes direct diagnosis difficult. Most attention has been given to sequence comparison between the hypervariable region of VP2 of classical isolates and of very virulent isolates. Sequence analysis of the vvIBDV isolate UK661 showed that only three unique (i.e., not found in non-vvIBDV isolates) amino acid substitutions are present within the hypervariable region of the VP2 protein. One amino acid substitution is present within the remaining part of the pVP2 protein, while 5 unique amino acid mutations are present within the VP4-encoding part of the polyprotein and 6 in the VP3-encoding part (Brown and Skinner, 1996). The smaller ORF of the UK661 isolate A-segment, encoding the VP5 protein, contains 2 unique amino acid substitutions. Additionally, 16 unique amino acid substitutions were found in the VP1 protein encoded by the B-segment of this vvIBDV isolate. The virulent phenotype of the vvIBDV might be influenced by each of the found amino acid substitutions, and even (silent) nucleotide substitutions within the coding or non-coding parts of either the A- or B-segment may contribute to the altered phenotype of the vvIBDV isolates in comparison with the classical or antigenic variant isolates. Serial passage on embryonated eggs of a vvIBDV isolate (OKYM) resulted in the appearance of a derivative isolate (OKYMT) which is able to grow on Chicken Embryo Fibroblast (CEF) cells and has lost its virulence. This adaptation was reported to be the result of 7 nucleotide substitutions in the polyprotein encoding part of the genome. Whether additional nucleotide substitutions (or deletions) were present in remaining parts of the A- or B-segment (e.g., untranslated regions, VP1-encoding region, and VP5-encoding region) was not determined (Yamaguchi et al., 1996). The reported nucleotide substitutions result in 5 amino acid substitutions. Three of these amino acid substitutions were located in the hydrophobic part of the hypervariable region (I256T, D279N, A284T) of VP2, one in the hydrophilic part located downstream of the hypervariable region (S315F) of VP2, and one in VP3 (A805T) (Yamaguchi et al., 1996). In an independent experiment, Yamaguchi et al. found that the adaptation of vvIBDV isolate TKSM into TKSMT resulted also in the A284T and D279N substitutions. The A284T substitution correlated in their analysis completely with adaptation onto CEF cells and loss of virulence. The D279N substitution was also present in both CEF-adapted vvIBDV isolates (OKYMT and TKSMT) and is potentially also important for growth on CEF cells and loss of virulence. The non-CEF adapted, classical IBDV isolate GBF-1 has, on the other hand, an asparagine at position 279, in combination with alanine at position 284, and cannot grow on CEF cells, so the single substitution D279N does not account for loss of virulence and growth on CEF cells. The amino acid changes in the VP2 apparently allow the modified IBDV to propagate on cells which do not have a receptor for wild-type IBDV. Cells possessing a wild-type IBDV receptor such as bursa cells are susceptible for classical and vvIBDV isolates. Recently, it was shown that amino acid substitution, A284T in combination with D279N, is indeed enough to turn a non-CEF-adapted very virulent IBDV isolate into a CEF-adapted isolate. Lim et al. introduced these two amino acid substitutions into the A-segment cDNA of vvIBDV isolate HK46 (Lim et al., 1999). After transfection of this cDNA, Lim et al. obtained an rIBDV isolate which possessed the phenotype of a CEF-culture-adapted isolate, i.e., an rIBDV isolate which is able to infect and multiply in CEF cells. The virulence of this rIBDV isolate was not assessed in chickens. Noteworthy, Lim et al. were unable to produce a recombinant infectious vvIBDV isolate using the unmodified cDNA of the HK46 isolate (Lim et al., 1999).
The goal of vaccination against IBD is prevention of subclinical and clinical IBD and the economic aspects of each. Effective vaccination for IBD can be divided into the following categories:
Protection of the developing bursa in broilers, breeders and layers.
Prevention of clinical disease in broilers, breeders and layers.
Priming and boosting of breeders.
To minimize the immunosuppressive effects of IBDV, the young chick must be protected. Protection of the very young can be achieved through high enough levels of maternal antibodies passed from the breeder hen to her progeny. Vaccination of the very young chick itself may not be successful since onset of protection after vaccination is between three and five days. When a bird, lacking maternal antibodies against IBDV, is exposed to a pathogenic IBDV field strain, damage will occur within 24–48 hours.
Generally, the early vaccinations of the breeders serve as priming. In most situations, this single vaccination is not considered to be adequate. “Boosting” is the term commonly associated with the administration of a final IBDV vaccination prior to the onset of lay. This is done to increase the circulating antibody in the hen and hence the maternal antibodies in the progeny. Both inactivated (oil emulsion vaccine) and live vaccines (IBDV) have been used for this purpose. The use of a live vaccine in an older bird will result in an increase of antibodies; however, large variations in antibody titers are often seen. These variations result in progeny becoming susceptible to field challenge from as early as a few days after hatching to 21 days after hatching. The use of inactivated IBDV vaccines gives a higher antibody titer as well as a decrease of variation between antibody titers of birds belonging to the same flock. The levels of maternal antibodies necessary to neutralize IBD vary with the invasiveness and pathogenicity of the field strain. In practical terms, if a very virulent IBDV isolate is present, higher maternal antibody levels are desired (see Table 1 for an overview of virulence of field isolates and strength of vaccines). Yet, for effective vaccination, avoiding interference with maternal antibodies is needed to induce a good immune response. Clinical IBD is typically seen between three and six weeks of age. The immune response of the chick must be stimulated as the passive protection is declining. The timing of the active vaccination may be estimated by the breeder or chick titer and the half-life of antibodies of approximately 3.5 days (De Wit and Van Loon, 1998; Kouwenhoven and Van den Bos, 1995). The levels of maternal antibodies tend to vary within a population. This variation might be a result of variation in the antibody levels of the breeder hen. Also, the mixing of progeny from several breeder flocks (e.g., combination of breeders of different age; breeders vaccinated with live vaccine and those with oil emulsion vaccine) results in variation of IBDV antibodies between chicks belonging to the same flock. If the coefficient of variation (CV) in mean maternal antibody titers is too wide, it may be recommended to vaccinate twice (with a 10 day interval) or to vaccinate early with a hot vaccine (in the presence of a high antigenic pressure).
The average titer of antibody against IBDV in a flock will decline in time (FIG. 1). As a result of the decrease in average antibody titers, an immunity gap will occur. The best results are obtained if the immunity gap is as short as possible and is as early as possible, within a minimum of 2 weeks after hatching. There should be at least sufficient immunity after active vaccination at the age of 4 weeks, since many handlings occur in the houses from that time point with risks of introducing field virus. Therefore, farmers like to vaccinate at 2 weeks or even before. Intermediate vaccines are often unable to break through the average IBDV antibody titer of the broiler at two weeks after hatching (FIG. 1). If there is a high variation in mean maternal antibody titers, some chicks will be effectively vaccinated with intermediate vaccines, others not. To circumvent those problems, hot vaccines are being used. A drawback of usage of hot vaccines is that the bursa of chickens with low-to-moderate maternal antibody titers will be (partly) damaged.
There is a wide variety of IBDV vaccines available. Important aspects in vaccination strategies are the ability of the virus to replicate in the face of maternal antibody (invasiveness of the vaccine) and the spectrum of antigenic content (including antigenic variants). The ability of a vaccine virus to replicate in the face of maternal antibodies allows live vaccines to be categorized into three main groups: mild, intermediate, and intermediate plus or hot vaccines (see Table 1).
The initial vaccines for IBD were derived from classical IBDV isolates. These vaccines were moderately pathogenic IBDV strains with low passage numbers in embryonated eggs. These were often used in breeder programs to induce high levels of circulating antibodies. However, when given to a young bird with moderate or low levels of maternal antibodies, these vaccines could cause extensive bursal atrophy resulting in immunosuppression. Mild vaccines were subsequently developed to be used in these young birds. The attenuation of classical IBDV was done in tissue culture systems. Traditionally, attenuated strains for vaccines are generated by adapting IBDV strains to chicken embryoblast (CEF) cells or other appropriate cells or cell lines through serial passages. These vaccines are not immunosuppressive even when used in birds having no maternal antibodies. However, moderate and high levels of antibodies easily neutralize them. As breeder programs developed (including the use of adjuvant, inactivated vaccines), higher levels of maternal antibodies were generated in progeny. This reduces the effectiveness of these mild vaccines.
Intermediate-strength vaccines were to overcome the inadequacies of the mild vaccines. Some of the intermediate-strength vaccines were developed by cloning a field isolate on chicken cell cultures. Intermediate-strength vaccines are capable of establishing immunity in birds with moderate levels of maternal antibodies. These vaccines will cause some bursal atrophy in birds without maternal antibodies, but are considered not immunosuppressive.
Hot (strong) or intermediate plus vaccines were developed after the first outbreaks with vvIBDV. These vvIBDV isolates could break through higher levels of maternal immunity than the vaccines that were on the market at that time. Vaccination with intermediate vaccines always came too late in situations with high infection pressure with vvIBDV. Hot vaccines consist of vvIBDV strains with low-to-moderate passage in embryonated eggs or bursa-derived IBDV of chickens infected with vvIBDV isolates. Adapting vvIBDV on cells traditionally used for the generation of vaccines in general fails, since either these cells are non-permissive for vvIBDV or, when adapted to the cells, the vvIBDV in question had lost its very virulent character, making it useless for hot or intermediate plus vaccines. Hot or intermediate plus vaccines are desirably able to circumvent maternal immunity at an earlier age than intermediate vaccines but spread more within a flock. If intermediate plus and hot vaccines are used in chickens with moderate-to-high levels of maternal antibodies, there is no negative side effect on the bursa (Kouwenhoven and Van den Bos, 1995). If these vaccines are used in chickens with low-to-moderate levels of maternal immunity, this causes depletion of lymphoid cells in the bursa and a severe depletion of peripheral blood-B cells is found (Ducatelle et al., 1995). Although a recovery of bursal function has been observed, these vaccines should be used with precautions.
Live vaccines must be given in a way in which the virus will preferably reach the bursa, where it will quickly multiply and induce an immune response. Possible routes for application of live vaccines include drinking water, spray, subcutaneous and in ovo. Inactivated IBD vaccines are used in broiler breeders. They differ in some of the same ways as live vaccines. Their efficacy depends upon the spectrum of antigens they contain. Injectable oil-emulsion products may be given subcutaneously or intramuscularly.
A continuous monitoring of the field situation using an integrated quality control scheme including serology can be a valuable tool for continuously adapting preventive vaccination programs to changing epidemiological conditions. Also, a continuous follow-up of the epidemiological situation will allow anticipation of the development of major epidemics (Ducatelle et al., 1995). However, the ability of diagnostic laboratories to monitor IBD with meaningful definitive data is difficult. Serology is important but can be confusing when all birds monitored from commercial broiler flocks have high levels of the same spectrum of circulating antibodies. Field evaluations of broilers to monitor the status of IBD are highly subjective: it is difficult to discriminate antibody titers obtained after vaccination from those induced by IBDV field infections. If it were possible to discriminate between IBDV antibody response to field virus and IBDV vaccination, it is possible to have “early warning” systems and to start IBDV eradication programs if desired. Only when there is a known difference between the antibody response to the used IBDV vaccine and IBDV field isolates, defined conclusion about whether (sub)clinical signs of IBDV are the result of live IBDV vaccination or of IBDV field isolates can be made.
The invention provides infectious recombinant Infectious Bursal Disease Virus (rIBDV) essentially incapable of growing in a cell that is not derived from a bursa cell or another cell comprising a wild-type IBDV receptor (a non-bursa cell). A bursa is a lymphoid organ, mostly comprising cells that are related to the immune system. In particular, it comprises lymphocytes or lymphocyte precursor cells of sometimes the T-cell but mainly the B-cell-type, and cells derived thereof, in close relation with monocytes or monocyte-derived cells such as macrophages, and also with follicular dendritic cells and antigen-presenting cells. In particular, the invention provides rIBDV that is essentially incapable of growing in a cell not listed among the above-identified bursa cells or cells derived thereof, such as dendritic cells, monocytes, lymphocytes or cells derived thereof. Herewith, the invention provides an rIBDV having retained an important characteristic, in that in comparison with commonly attenuated IBDV strains, it cannot or only little grow in non-bursa cells, such as the well-known CEF, QM5 or VERO cells, or other cells that are commonly used for propagating attenuated strains of IBDV. In particular, the invention provides an rIBDV essentially incapable of growing in a non-B-cell-derived cell. Essentially incapable of growing herein means that the isolate in question is not or only little capable to infect, multiply or be released for further replication. No such rIBDV isolates existed prior to this invention; all previously known rIBDV isolates grow in non-bursa cell-derived cells such as CEF cells (WO98/09646; Lim et al., 1999), thereby, for example, having lost those very virulent characteristics essential for maintaining in a vaccine strain designed to face the above-identified problems.
In a preferred embodiment, the invention provides are infectious rIBDV having retained at least part of the very virulent characteristics of a very virulent Infectious Bursal Disease Virus (vvIBDV) needed to provide protection against vvIBDV. In particular, vvIBDV is provided that is essentially incapable of growing in a non-bursa cell-derived cell. In particular, as, for example, demonstrated in the detailed description, the invention provides an rIBDV essentially incapable of growing in a CEF cell, a VERO cell or a QM5 cell, except, of course, in those CEF, VERO, QM5, or related cells having been provided with the necessary means (such as a transgenic receptor or replication system derived from a bursa cell or other characteristic of B-lymphoid cells which support the propagation of IBDV in an essentially non-lymphoid cell if this characteristic is present in such a non-B-lymphoid cell) needed for replication of classical or very virulent IBDV.
Furthermore, the invention provides an rIBDV wherein the amino acid sequence of protein VP2 comprises no asparagine at amino acid position 279, but, for example, comprises an amino acid particular for a strain with a very virulent character, such as with aspartic acid at amino acid position 279. Such rIBDV strains as provided by the invention have retained at least part of the very virulent characteristics of vvIBDV, as well as an rIBDV according to the invention wherein the amino acid sequence of protein VP2 comprises no threonine at amino acid position 284, but, for example, comprises an amino acid particular for a strain with a very virulent character, such as with alanine at amino acid position 284.
In a preferred embodiment, the invention provides an rIBDV according to the invention wherein the amino acid sequence of protein VP2 at least comprises a stretch of amino acids from about position 279 to 289, preferably from about position 229 to 314, most preferably from about position 214 to 328 as found in a vvIBDV isolate such as shown in Table 8.
The invention furthermore provides a method for obtaining an infectious recombinant copy of Infectious Bursal Disease Virus (rIBDV) essentially incapable of growing in a non-bursa cell-derived cell or having at least part of the very virulent characteristics of a very virulent Infectious Bursal Disease Virus (vvIBDV) comprising transfecting at least one first cell with a nucleic acid such as a cDNA or RNA comprising an IBDV genome at least partly derived from a vvIBDV, incubating the first cell in a culture medium, recovering rIBDV from the transfected first cell or culture medium, and propagating recovered rIBDV in at least one second cell which is permissive for vvIBDV. A vaccine derived from the recombinant virus as described is also part of this invention. Also, a vaccine comprising a chemically or physically inactivated recombinant virus or parts thereof is part of this invention.
Also, the attenuated derivatives of initially produced recombinant very virulent IBDV are part of this invention. Such a virus can be attenuated by known methods including serial passage, removing specific nucleic acid sequences, or site-directed mutagenesis. Physiologically acceptable carriers for vaccines of poultry are known in the art and need not be further described herein. Other additives, such as adjuvants and stabilizers, among others, may also be contained in the vaccine in amounts known in the art. Preferably, adjuvants such as aluminum hydroxide, aluminum phosphate, plant and animal oils, and the like, are administered with the vaccine in amounts sufficient to enhance the immune response to the IBDV. The vaccine of the present invention may also contain various stabilizers. Any suitable stabilizer can be used including carbohydrates such as sorbitol, mannitol, starch, sucrose, dextrin, or glucose; proteins such as albumin or casein; and buffers such as alkaline metal phosphate and the like. A stabilizer is particularly advantageous when a dry vaccine preparation is prepared by lyophilization. The vaccine can be administered by any suitable known method of inoculating poultry including nasally, ophthalmically, by injection, in drinking water, in the feed, by exposure, and the like. Preferably, the vaccine is administered by mass administration techniques such as in ovo vaccination, by placing the vaccine in drinking water or by spraying the animals' environment. When administered by injection, the vaccines are preferably administered parenterally. The vaccine of the present invention is administered to poultry to prevent IBD anytime before or after hatching. Poultry is defined to include but not be limited to chickens, roosters, hens, broilers, roasters, breeders, layers, turkeys and ducks. Examples of pharmaceutically acceptable carriers are diluents and inert pharmaceutical carriers known in the art. Preferably, the carrier or diluent is one compatible with the administration of the vaccine by mass administration techniques. However, the carrier or diluent may also be compatible with other administration methods such as injection, eye drops, nose drops, and the like.
As explained above, there is need for an IBDV vaccine that can protect against field infections with IBDV, and preferably against very virulent IBDV (vvIBDV). It is clear that vaccines derived from attenuated classical strains and not from very virulent strains will not be able to sufficiently protect. However, as explained above, simply adapting and cultivating a vvIBDV strain on a cell or cell-line, such as VERO, CEF or QM5, as one skilled in the art would first do in order to obtain a vaccine strain from a vvIBDV strain, reduces it virulent phenotype such that no sufficient protection is to be expected. Therefore, a vaccine strain is needed that has at least partly maintained the very virulent or hot character, in order to provide sufficient protection; however, paradoxically, such a desirable vaccine strain would most likely not be able to be sufficiently or substantially propagated on appropriate cells, such as non-bursa cell-derived VERO, CEF or QM5, deemed needed to obtain the vaccine. In a preferred embodiment, the invention provides a method wherein the first cell is a non-bursa cell-derived cell non-permissive for vvIBDV, preferably wherein the first cell has additionally been provided with a helper virus or a viral protein (herein T7-polymerase is used) derived thereof. With the help of such a cell comprising a properly selected helper virus, e.g., expressing distinct IBDV or Birna virus viral proteins, or of a cell expressing IBDV or Birna virus viral proteins, (also called a complementary cell) also now defective or deficient rIBDV can be made.
The invention therewith also provides a method to generate Infectious Bursal Disease Virus, by combining cDNA sequences derived from very virulent IBDV (vvIBDV) isolates with cDNA sequences derived from either serotype I classical attenuated IBDV isolates, serotype I antigenic variants of IBDV, or serotype II IBDV isolates, wherein the infectious copy recombinant Infectious Bursal Disease Virus, having retained at least part of the very virulent characteristics of a very virulent Infectious Bursal Disease Virus, has at least retained the incapacity to substantially be propagated on a vvIBDV non-permissive cell such as a QM5 or CEF cell.
Preferably, a method as provided by the invention provides a vaccine comprising an IBDV genome wherein parts of segments A and/or B derived from a vvIBDV are used combined with parts of segments A and/or B derived from an attenuated IBDV, such as attenuated serotype I or I IBDV. Such an rIBDV is herein also called a mosaic IBDV (mIBDV). Herein we show that (mosaic) vvIBDV can be generated from cDNA by transfection of non-susceptible cells followed by amplification of the cDNA-derived rIBDV on susceptible cells. The invention provided herein provides a method to generate vvIBDV from the cloned, full-length cDNA of a vvIBDV isolate (see Table 5 and 6). After transfection of QM5 cells with cDNA of vvIBDV, it is essential that propagation of the generated vvIBDV virus takes place on cells which are permissive for vvIBDV. These permissive cells can, for example, be found among bursa cell-derived cells such as primary bursa cells, in chicken in embryo cells, chicken embryo's, or young chickens. Using the method described herein we have, for example, produced recombinant D6948 (rD6948) using the cDNA derived from the very virulent D6948 IBDV isolate. This rD6948 isolate has the same virulence as the parental D6948 isolate (Table 6).
Preferably, the invention provides a method wherein a permissive second cell is a primary bursa cell, thereby allowing initial propagation of the desired vaccine virus. As explained above, there is a need for a vaccine capable of breaking through maternal immunity of young chickens at an early stage. A desired vaccine should preferably be able to induce a high level of protection in vaccinated young chickens, and should, therefore, be as immunogenic as very virulent viruses or be almost as immunogenic.
The invention furthermore provides a method to engineer recombinant mosaic IBDV (mIBDV) vaccine which has one or more of the desired phenotypes, i.e., i) being able to break through high levels of maternal antibodies in young chickens and being highly immunogenic, and ii) having a reduced pathogenicity compared to wild-type very virulent IBDV isolates. In one embodiment, the invention provides an infectious mosaic IBDV (mIBDV) comprising an rIBDV wherein at least one genome segment comprises nucleic acid derived from at least two different Birna virus isolates. It is preferred that at least one of the isolates is a vvIBDV characterized by its incapacity to substantially be propagated on a vvIBDV non-permissive cell, such as a VERO, QM5 or CEF cell and/or characterized by its capacity to substantially be propagated on a vvIBDV-permissive cell, such as a primary bursa cell. For example, the invention provides mIBDV which consists partly of the genome derived from a classical attenuated isolate (such as CEF94) and partly derived from the genome of a vvIBDV isolate (such as D6948). A recombinant mosaic IBDV (mIBDV), made on the basis of infectious cDNA derived from a very virulent IBDV isolate (D6948) and combined with defined parts of cDNA derived from a cell culture adapted, serotype I, classical IBDV isolate (CEF94), results in an mIBDV isolate which has a reduced pathogenicity compared to wild-type vvIBDV isolates which are essentially not replicating on non-B-lymphoid-derived cells.
Furthermore, specific nucleotide substitutions which either do or do not lead to amino acid mutations, or deletion of specific parts of the IBDV genome, again lead to an altered phenotype of the generated mIBDV. For example, the replacement of the pVP2-coding region of CEF-94 cDNA with the corresponding region of cDNA of D6948 yielded plasmid pHB36-vvVP2. This plasmid was subsequently transfected into FPT7-infected (Britton et al., 1996) QM5 cells in combination with pHB-34Z. Supernatant of these transfected QM5 cells was subsequently transferred to fresh QM5 cells. None of these QM5 cells reacted positively in an IPMA using specific antibodies for the VP3 protein of IBDV. On the other hand, primary bursa cells, after being overlaid with supernatant of the transfected cells, reacted positively in the same IPMA. The functional feature of being able to enter permissive cells such as QM5 cells is apparently located in the pVP2 coding region of the A-segment. This invention provides a method to generate recombinant vvIBDV (such as rD6948) having a pVP2 sequence exactly as found in a wild-type vvIBDV (here D6948). All very virulent isolates of which the pVP2 sequences have been described thus far have an alanine at position 284 and cannot or only little be propagated on CEF cells (see Tables 7 and 8). On the other hand, when a threonine is present at position 284, propagation on CEF cells is possible, but this is associated with the lack of a very virulent phenotype (see Tables 7 and 8). Herein we describe a method to generate infectious recombinant IBDV (rIBDV) having the nucleotide sequence of a wild-type very virulent IBDV isolate, including the alanine codon for amino acid 284 and being unable to be propagated on CEF cells. Furthermore, in our rD6948 isolate we have an aspartic acid present at position 279 instead of an asparagine commonly found for virulent IBDV isolates which can be propagated on CEF cells (Tables 7 and 8). The rD6948 is truly a very virulent rIBDV, as it is unable to grow on CEF cells (Table 5) and induces similar clinical signs and mortality as wild-type very virulent D6948 isolate (Table 6). Although mIBDV isolate (mCEF94-vvVP2) did not cause any mortality or body weight loss in a 9-day course, in contrast to the D6948, rD6948 and srIBDV-DACB isolates (also having a functional VP2 protein derived from vvIBDV, see Table 6), it caused the same reduction in bursa weight after 9 days post-infection as the wild-type very virulent D6948 isolate.
In yet another embodiment, the invention provided a mosaic IBDV according to the invention wherein at least one of the isolates is a serotype II IBDV. Such an mIBDV, preferably lacking at least one immunodominant epitope specific for a serotype I IBDV as well, is an (r)D6948-derived vaccine virus such as mD6948-s2VP3C1, also having a functional VP2 protein derived from vvIBDV, allowing vaccination with a marker vaccine. Vaccination with an IBDV marker vaccine and subsequent testing with a corresponding diagnostic test enables the discrimination between antibodies induced by the vaccine and by infection with IBDV field isolates. This mIBDV can be differentiated from all other known wild-type IBDV isolates, either belonging to serotype I or serotype II, for example, by using a specific set of monoclonal antibodies. The generation of mIBDV from serotype I and II cDNA provides such an mIBDV marker vaccine that induces a serological response in chickens that can be differentiated from the serological response induced by IBDV field strains. The marker vaccine provided by the invention, lacking at least one immunodominant epitope, preferably a serotype I epitope, enables the discrimination between vaccinated and infected animals by means of a diagnostic serologic test. Such an mIBDV marker vaccine is preferably based upon vvIBDV and contains specific sequences originating from classical serotype I or serotype II IBDV. Such an mIBDV marker vaccine has one or more of the following characteristics: i) It induces a protective immune response against vvIBDV field viruses despite high levels of maternal antibodies. ii) It has a reduced pathogenicity compared to vaccines based upon wild-type vvIBDV. iii) It, for example, misses at least one serotype I-specific antigen which enables the serological discrimination of the mIBDV marker vaccine from all serotype I IBDV isolates.
Also, the invention provides a method to produce or generate tailor-made vaccines against specific antigenic variants of IBDV by incorporating the specific amino acid changes in an mIBDV vaccine virus. Depending on the composition, these mosaic IBD viruses (mIBDV) possess different phenotypes and different antigenic properties. A specific mutation in one of the viral proteins can have a profound effect on IBDV viability. We found that this is true in the case of a single nucleotide substitution, leading to a single amino acid mutation in VP4 (V582A). No rIBDV could be rescued from cDNA when this particular nucleotide substitution was present. Not only mutations within the VP4-encoding region itself, but also mutations or deletions in the region of the cleavage sites (pVP2-VP4 and VP4-VP3) may have a negative effect on replication of rIBDV. Mutations in the other viral proteins, or even deletion of an entire viral protein (i.e., VP5), influence the replication and or virulence as well. Two groups have constructed a VP5 minus rIBDV isolate, by introducing mutations in the cDNA of a CEF-adapted D78 IBDV isolate (Mundt et al., 1997; Yao et al., 1998). Apparently, the VP5 protein, which is a non-structural protein, is also a non-essential protein. Yao et al. reported that inactivation of the ORF for VP5 (replacement of the start codon by a stop codon) yielded infectious rIBDV (rD78NS which grows to slightly lower titers (in vitro) than rD78), while Mundt et al. reported that inactivation of the ORF for VP5 (replacement of the start codon by an arginine codon) yielded an rIBDV (IBDV/VP5−) which is able to grow to the same titers (in vitro) as the parental isolate. Furthermore, Yao et al. reported that rD78NS has decreased cytotoxic and apoptotic effects in cell culture (in vitro) and has a delay in replication compared to the parental isolate (in vivo), and failed to induce any pathological lesions or clinical signs of disease in infected chickens.
Mutations or deletions in the mIBDV cDNA yields an mIBDV with a desired phenotype, i.e., mIBDV which is based on a very virulent isolate but which has a reduced ability to replicate and hence a reduced pathogenicity. The introduction of cDNA sequences from a serotype II, cell culture adapted, IBDV isolate (TY89) into the mosaic virus gives us yet another opportunity to generate marker mIBDV vaccine which can be discriminated from wild-type serotype I IBDV, for example, by using specific monoclonal antibodies. Such mIBDV can be used to induce an antibody spectrum, which differs from the spectrum induced by IBDV field isolates. This enables the development of a serologic test to determine whether IBDV antibodies are the result of live mIBDV vaccination or of infection with IBDV field isolates. For example, the mCEF94-s2VP3C virus is recognized by serotype II-specific VP3 antibody (Mab T75), while it is also recognized by a serotype I-specific VP2 antibody (Mab 1.4). This particular rIBDV is, on the other hand, not recognized by a serotype I-specific VP3 antibody (Mab B10). No apparent difference is present between the replication of mCEF94-s2VP3C and rCEF94, indicating that the exchange of the VP3C-terminal part does not lead to major changes in replication ability in QM5 cells. When, on the other hand, the complete VP3-encoding region was exchanged, we observed a severe reduction in replication ability of the resulting virus (mCEF94-s2VP3). On the other hand, mCEF94-s2VP3N was not reacting with Mab C3 (VP3, serotype I) while it is fully reacting with Mab B10 (VP3, serotype I) and only partially with Mab T75 (VP3, serotype II. Replication of this mosaic IBDV on CEF cells is reduced compared to rCEF94. From the generated mIBDV, based on cDNA derived from serotype I (CEF94) and serotype II (TY89), it is clear that a serological marker based on VP3 has been identified. The replacement of the cDNA of (part of) VP3 of serotype I for the corresponding part of serotype II leads to a unique combination of IBDV antigens within one mIBDV isolate. An mIBDV isolate based on this combination of antigens can be used as an IBDV marker vaccine.
The introduction of the VP3 C-terminal part of TY89 (Serotype II) into the cDNA of D6948 yielded a mosaic IBDV (mD6948-s2VP3C1) which has a reduced virulence (no mortality, no body weight loss) compared to D6948 or rD6948 (Table 6). This mIBDV, or a comparable isolate which is more or less virulent, is also advantageously used as an IBDV marker vaccine to prevent infections with very virulent IBDV field isolates.
Furthermore, the invention provides using site-specific mutagenesis techniques to introduce any desired nucleotide mutation within the entire genome of mIBDV. Using this technique allows adapting mIBDV vaccines to future antigenic variations by including any mutation that has been found in antigenic variant IBDV field isolates. Furthermore, it is provided by the invention to exchange part of the genomic sequence of IBDV with the corresponding part of a Birna virus belonging to another genus (e.g., DXV, EPNV, OV, TV). Herewith, the invention provides new mosaic Birna (mBirna) viruses which have new characteristics resulting in new recombinant vaccines for IBDV or other Birna viruses. Also, the use of cDNA of other Birna viruses (e.g., DXV, IPNV, OV or TV) leads to new IBDV vaccines. In this approach, one or more of the IBDV immunodominant or neutralizing epitopes are exchanged with the corresponding parts of the protein of another Birna virus.
Of course, the invention also provides a method for producing an rIBDV according to the invention, the vector comprising heterologous nucleic acid sequences derived from another virus, or (micro)organism, whereby r- or mIBDV serves as a vector. For example, a method is provided to generate an infectious copy IBDV which expresses one or more antigens from other pathogens and which can be used to vaccinate against multiple diseases. Such an infectious copy IBDV, for example, comprises a heterologous cDNA encoding a heterologous protein obtained from a pathogen, for example, poultry pathogens. Also, a method is provided to generate a conditional lethal IBDV deletion mutant which can be used as a self-restricted, non-transmissible (carrier) vaccine. Such an IBDV deletion mutant is unable to express one of the IBDV proteins, and is phenotypically complemented by supplying the missing protein by other means.
The invention is further explained in the detailed description without limiting the invention thereto.